Article Text

Download PDFPDF

Research paper
Pathophysiological insights into ALS with C9ORF72 expansions
  1. Kelly L Williams1,2,3,
  2. Jennifer A Fifita1,2,
  3. Steve Vucic4,
  4. Jennifer C Durnall2,
  5. Matthew C Kiernan5,
  6. Ian P Blair1,2,3,
  7. Garth A Nicholson2,3,6
  1. 1Australian School of Advanced Medicine, Macquarie University, Sydney, New South Wales, Australia
  2. 2Northcott Neuroscience Laboratory, ANZAC Research Institute, Sydney, New South Wales, Australia
  3. 3Sydney Medical School, University of Sydney, Sydney, New South Wales, Australia
  4. 4Department of Neurology, Westmead Hospital, Sydney, New South Wales, Australia
  5. 5Prince of Wales Medical, Research Institute and Prince of Wales Clinical School, University of New South Wales, Sydney, New South Wales, Australia
  6. 6Molecular Medicine Laboratory, Concord Hospital, Concord, New South Wales, Australia
  1. Correspondence to Dr Ian P Blair, Australian School of Advanced Medicine, Macquarie University, 2 Technology Place, Sydney, NSW 2109, Australia; ian.blair{at}


Objective Expansions of a hexanucleotide repeat in C9ORF72 are a common cause of familial amyotrophic lateral sclerosis (ALS) and a small proportion of sporadic ALS cases. We sought to examine clinical and neurophysiological features of familial and sporadic ALS with C9ORF72 expansions.

Methods C9ORF72 was screened for expansions in familial and sporadic ALS. Clinical features of expansion positive cases are described. Cortical excitability studies used novel threshold tracking transcranal magnetic stimulation techniques with motor evoked responses recorded over the abductor pollicis brevis.

Results and conclusions Analysis of large clinical cohorts identified C9ORF72 expansions in 38.5% (72/187) of ALS families and 3.5% (21/606) of sporadic ALS cases. Two expansion positive families were known to carry reported ANG mutations, possibly implicating an oligogenic model of ALS. 6% of familial ALS cases with C9ORF72 expansions were also diagnosed with dementia. The penetrance of ALS was 50% at age 58 years in male subjects and 63 years in female subjects. 100% penetrance of ALS was observed in male subjects by 86 years, while 6% of female subjects remained asymptomatic at age 82 years. Gender specific differences in age of onset were evident, with male subjects significantly more likely to develop ALS at a younger age. Importantly, features of cortical hyperexcitability were apparent in C9ORF72-linked familial ALS as demonstrated by significant reduction in short interval intracortical inhibition and cortical silent period duration along with an increase in intracortical facilitation and motor evoked potential amplitude, indicating that cortical hyperexcitability is an intrinsic process in C9ORF72-linked ALS.

  • ALS
  • Dementia
  • Neurogenetics

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.


Amyotrophic lateral sclerosis (ALS) is a fatal neurodegenerative disease characterised by the progressive loss of upper and lower motor neurones. Inheritance of the disease is seen in approximately 10% of ALS, with the remainder occurring as apparently sporadic cases. Comorbidity with frontotemporal dementia (FTD) is seen in around 20% of ALS cases, and as many as 50% show evidence of frontal lobe dysfunction.1 Around 25%–35% of familial ALS cases have been attributed to mutations in genes including SOD1, TARDBP, FUS and UBQLN2.2–5 Rare mutations in OPTN, VCP and FIG4 are also thought to account for a small proportion of cases (reviewed in Andersen and Al-Chalabi6).

Pathogenic expansions of a non-coding hexanucleotide repeat sequence (GGGGCC) in the C9ORF72 gene were recently reported in familial and sporadic forms of ALS.7 ,8 The C9ORF72 ALS phenotype is characterised by an earlier age of onset, shorter survival and greater incidence of dementia and psychosis.9–11 Further, the penetrance of C9ORF72 expansions seems incomplete and variable, with some family members developing FTD, others ALS or a combination of FTD–ALS, while some family members fail to develop ALS.9–12 Given such a varied phenotype, the mechanism by which expanded hexanucleotide repeats in C9ORF72 lead to ALS remains to be clarified and a matter of ongoing research.

Of relevance, cortical hyperexcitability appears to precede the clinical development of SOD1 familial ALS, and has been linked to peripheral neurodegeneration.13 ,14 These findings suggest that SOD1-linked familial ALS may begin centrally, with motor neurone degeneration possibly mediated by an anterograde transsynaptic mechanism. If cortical hyperexcitability is confirmed in C9ORF72 familial ALS, this could shed further light on pathophysiological mechanisms in ALS.

Consequently, the aims of the present study were to determine the frequency and penetrance of C9ORF72 expansions in large familial and sporadic ALS cohorts, and examine the clinical phenotype and pathophysiological mechanisms, including whether cortical hyperexcitability is a feature of C9ORF72-linked familial ALS.

Materials and methods


A total of 193 familial ALS cases from 152 ALS families and a further 170 non-ALS family members were recruited through neurogenetic clinics at Concord Hospital, Sydney, as well as at the Molecular Medicine Laboratory, Concord Hospital, a referral centre for ALS DNA diagnostic testing. Most families were of European descent. DNA from 559 sporadic ALS cases were obtained from the Australian motor neuron disease (MND) DNA Bank, and a further 47 sporadic cases were ascertained through the neurogenetic clinics at Concord Hospital. All patients were diagnosed with definite or probable ALS according to El Escorial criteria.15 Patients and family members provided informed written consent regulated by the Human Research Ethics Committee of the Sydney South West Area Health Service. All families had previously been screened for mutations in known ALS genes including TARDBP, SOD1, FUS, OPTN, VCP, UBQLN2, ANG, FIG4, DCTN1 and CHMP2B.

Genetic analysis

Genomic DNA was extracted from peripheral blood using standard protocols. Analysis of the C9ORF72 hexanucleotide repeat and expansion was performed using the method described in Renton et al.8 Fragment length analysis was performed on an ABI 3730XL DNA Analyzer (Applied Biosystems) and data were analysed using GeneMarker software V.1.95 (SoftGenetics, Pennsylvania, USA).

Statistical analysis

Statistical analysis was performed using MedCalc V.12.3.0 (Mariakerke, Belgium). All individuals with an expansion, including affected and asymptomatic carriers, were used to calculate the cumulative probability of disease onset by a particular age using the Kaplan–Meier method.16 Differences in age of onset between male and female subjects were compared using the log-rank test. A p value less than 0.05 was considered significant. To minimise ascertainment bias, sporadic cases were excluded when expansion results were not available for both parents.

Cortical excitability testing

Cortical excitability was assessed by applying transcranial magnetic stimulation to the motor cortex by means of a 90 mm circular coil, with currents generated by two high-power magnetic stimulators which were connected via a BiStim device (Magstim Co., Whitland, UK). A paired-pulse threshold tracking paradigm was used to determine the resting motor threshold (RMT), short interval intracortical inhibition and intracortical facilitation according to a previously described protocol.17 In addition, single stimulus transcranial magnetic stimulation was used to record the maximal motor evoked potential (MEP) response and cortical silent period duration with the stimulus intensity set to 140% of RMT. Central motor conduction time was recorded using the F-wave method.18

The compound muscle action potential (CMAP) and neurophysiological index were recorded for the target abductor pollicis brevis muscle as previously described.17 ,19


Genetic analysis

Repeat-primed PCR analysis of a clinically affected index case from each Australian ALS family established that an expansion of the C9ORF72 hexanucleotide repeat was present in 72 of 187 (38.5%) families. Segregation of the C9ORF72 hexanucleotide repeat expansion was further confirmed in a second affected family member in 15 families. Across all expansion positive ALS families there were 65 (46 female, 19 male subjects) asymptomatic mutation carriers with a mean age of 56.5 years (range 16–77 years). Kaplan–Meier analysis of the age of ALS onset among expansion positive individuals demonstrated a significant difference between male and female subjects (p=0.0173), with male subjects more likely to develop ALS at a younger age (figure 1). Penetrance was less than 10% in male subjects under 47 years and female subjects under 48 years, 50% at 58 years in male subjects and 63 years in female subjects, and 100% in male subjects by 86 years. Female subjects reached 94% penetrance by 82 years.

Figure 1

Kaplan–Meier curves of the cumulative incidence of amyotrophic lateral sclerosis (ALS) among 213 subjects carrying a C9ORF72 hexanucleotide repeat expansion. This provides a likelihood estimate that an individual who carries the repeat expansion will be affected by a certain age.

Repeat-primed PCR analysis identified a C9ORF72 expansion in 21 of 606 (3.5%) of sporadic ALS cases in our cohort. We note that 263 of the 606 Australian sporadic cases analysed here were previously included in the study performed in Majounie et al.12

Clinical features of ALS patients with C9ORF72 expansion

Table 1 summarises the clinical data that were available for ALS patients in the 72 families and the 21 sporadic cases with C9ORF72 repeat expansions. In short, there were no significant clinical differences between familial and sporadic ALS. The number of male subjects compared with female subjects with the expansion was similar. The mean age of onset was slightly lower in familial ALS compared to sporadic ALS, whereas the mean duration of the disease was similar. The range of disease onset was much greater in familial than sporadic ALS (20–86 years compared with 46–75 years, respectively), as was the range of disease duration (0.5–16 years compared with 1–4 years, respectively). The proportion of bulbar onset cases to spinal onset cases was similar in familial and sporadic ALS. ALS-dementia was evident in 4.8%, and dementia alone (without ALS) in 3.3% of all expansion positive individuals.

Table 1

Summary of clinical data from 72 families and the 21 sporadic cases with C9ORF72 repeat expansions

Index cases from two families with repeat expansions were previously shown by us to carry reported ANG mutations (p.K17I and p.I46V, data not shown).20 The p.I46V variant was present in two expansion positive ALS patients from the same family. The ages of onset seen in these two patients were 51 and 58 years and were not significantly different to the average age of onset seen in the overall cohort (57.6 years). The age of onset seen in a patient with a C9ORF72 expansion and ANG p.K17I mutation was 76 years. No unusual clinical features were evident in these three patients and disease durations were not statistically different from that of the overall cohort.

Neurophysiological assessment

In order to glean further insights into the pathophysiological mechanisms underlying C9ORF72 familial ALS, cortical excitability testing was undertaken. Prior to undertaking cortical excitability studies, however, the peripheral disease burden was assessed. CMAP amplitude (ALSC9ORF72 5.0±1.1 mV; ALSSPORADIC 5.6±0.3 mV; 95% CI 9.8 to 11.4 mV) and neurophysiological index (ALSC9ORF72 0.9±0.7; ALSSPORADIC 0.7±0.1; 95% CI 2.4 to 2.8) were significantly reduced in C9ORF72 familial ALS patients when compared with controls but similar to sporadic ALS, thereby suggesting that the peripheral disease burden was similar between C9ORF72 familial ALS and sporadic ALS.

Assessment of cortical excitability

Cortical excitability studies were undertaken in five patients. Short interval intracortical inhibition, defined as an increase in the test stimulus intensity required to track a constant target MEP response of 0.2 mV, was significantly reduced in C9ORF72 familial ALS patients when compared with controls (ALS C9ORF72−0.2±2.4%; 95% CI 9.1% to 12.0%, figure 2A) but similar when compared with sporadic ALS patients (2.6±0.9%, p=0.18). Further, intracortical facilitation was increased in the C9ORF72 familial ALS patients when compared with controls (ALSC9ORF72 −3.7±1.8%; 95% CI −2.3% to 0.7%, figure 2A) but again similar to sporadic ALS (−2.2±0.6%). Of further relevance, RMT, defined as the unconditioned stimulus intensity required to produce and maintain the target MEP response, was reduced in the C9ORF72 familial ALS patients (ALSC9ORF72 52.1±3.4%; 95% CI 57.0% to 61.1%). Importantly, there was no significant difference of RMT between C9ORF72 familial and sporadic patients (ALSSPORADIC 57.2±0.9%).

Figure 2

Assessment of cortical excitability in amyotrophic lateral sclerosis (ALS) patients with C9ORF72 mutations. (A) Short interval intracortical inhibition was reduced, while intracortical facilitation increased in ALS patients with C9ORF72 mutations when compared with controls. (B) Motor evoked potential (MEP) amplitude, expressed as a percentage of the peripheral compound muscle action potential (CMAP) response, was increased in ALS patients with C9ORF72 mutations and sporadic ALS when compared with controls.

The MEP amplitude, expressed as a percentage of the CMAP amplitude, was significantly increased in the C9ORF72-linked familial ALS patients when compared with controls (ALSC9ORF72 45.7±4.5%, 95% CI 22.6% to 31.4%), but comparable with sporadic ALS (38.1±2.2%, figure 2B). In addition, the cortical silent period was significantly reduced in C9ORF72-linked familial ALS and sporadic ALS patients (ALSC9ORF72 196.7±3.5 ms; ALSSPORADIC 181.4±4.3 ms; 95% CI 201.9 to 214.4 ms). In contrast, central motor conduction time was similar in the three groups (ALSC9ORF72, 4.8±0.2 ms; sporadic ALS, 5.1±0.2 ms; controls, 5.1±0.2 ms).


The present study assessed the presence of C9ORF72 hexanucleotide repeat expansions in large familial and sporadic ALS cohorts, further expands the phenotype associated with this form of ALS and suggests a potential pathophysiological mechanism, namely, cortical hyperexcitability.

C9ORF72 repeat expansions were found in 38.5% of Australian ALS families and 3.5% of sporadic ALS. Both of these cohorts are of predominantly European ancestry. This repeat frequency for familial ALS is similar to that previously reported in a combined cohort of white individuals from the USA, Europe and Australia (37.6%) but lower than that reported for sporadic ALS (7%).12

C9ORF72-linked ALS was highly penetrant in families, with 100% penetrance by 86 years in male subjects and 94% penetrance by 82 years in female subjects. However, gender specific differences were observed for age of onset. Around 71% (46/65) of carriers (no ALS or FTD) from Australian ALS families were female. Furthermore, Kaplan–Meier curves of age-dependent penetrance for affected individuals established that male subjects were more likely to express the disease at a younger age.

In all, 6% of familial ALS cases were diagnosed with dementia. No sporadic ALS cases were diagnosed with dementia. Previously reported frequencies of comorbid ALS-dementia vary widely from 7% to 86%.21 ,22 The frequency of ALS-dementia in the present cohort may be underestimated, with cases historically ascertained on the basis of ALS alone before the relationship between ALS and FTD was widely recognised.

Psychosis was previously reported in some ALS–FTD cases with the C9ORF72 repeat expansion.10 ,23 ,24 We identified two expansion positive ALS patients with unspecified dementia and psychosis. In addition, mood disorders were evident in some cases, with three patients diagnosed with depression, another with bipolar disorder, and one ALS family with a history of suicide. Although not common, psychosis and mood disorders may be considered part of the C9ORF72-linked phenotypic spectrum.

C9ORF72 expansions were identified in three ALS patients (from two families) that carried angiogenin p.K17I and P.I46V variants. The ANG p.K17I variant was originally reported in two sporadic ALS cases20 and subsequently shown to segregate with disease in five affected individuals from a family with ALS, FTD and Parkinsonism.25 The ANG p.I46V variant was initially described in one sporadic and two unrelated familial ALS patients20 and we have found segregation of this variant in two related ALS cases. Cooper-Knock et al26 and van Blitterswijk et al27 also found SOD1, TARDBP, FUS and OPTN variants coexisting with C9ORF72 expansions in several ALS patients. Earlier age of ALS onset and longer disease duration were reported for an expansion patient with a TARDBP p.A321V mutation.26 van Blitterswijk et al27 suggested that the presence of multiple ALS mutations in patients may provide evidence for an oligogenic model of ALS and this may explain non-penetrance in ALS as well as the phenotypic variability that is seen in C9ORF72 expansion patients and families. In the current study, ages of onset and disease durations were not significantly different for three patients who carried both a C9ORF72 expansion and a reported ANG mutation suggesting that the presence of two reported mutations does not appear to contribute to disease severity in these cases.

Pathophysiological mechanisms underlying C9ORF72 familial ALS

The detailed pathophysiological mechanisms underlying the development of C9ORF72-linked ALS remains to be fully elucidated. Previously, cortical hyperexcitability was reported to be an intrinsic and early feature of both sporadic ALS and SOD1-linked familial ALS.13 ,28 In addition, the clinical development of SOD1-linked familial ALS may be preceded by the development of cortical hyperexcitability,13 thereby suggesting that glutamate excitotoxicity underlies the development of motor neurone degeneration in ALS. In addition, degeneration of inhibitory cortical interneurones, as established in sporadic ALS,29 could also result in cortical hyperexcitability by disinhibiting the motor cortex. The findings in the present study establish that cortical hyperexcitability is a feature of C9ORF72-linked familial ALS, suggesting that glutamate excitotoxicity may be underlying the pathophysiological process. Given that the development of glutamate excitotoxicity has been related to astrocyte dysfunction,30–36 and that genetically modified astrocytes exert neuroprotective effects in motor neurone cell cultures,37 therapeutic approaches using cell based therapies, in particular modified astrocytes, may prove useful in the treatment of ALS. Longitudinal cortical excitability studies of presymptomatic C9ORF72 expansion carriers, along with assessment of gene carriers manifesting depression prior to ALS onset, may provide further insights into ALS pathophysiology.


We thank patients and family members for their ongoing participation and support. We also thank C Cecere and M Edwards for assistance with family recruitment and the Australian MND DNA Bank for assistance in sample collection.



  • KLW, JAF, IPB and GAN contributed equally.

  • Contributors KLW: drafting/revising the manuscript, study concept or design, acquisition of data, analysis or interpretation of data, statistical analysis. JAF, SV and MCK: drafting/revising the manuscript, study concept or design, acquisition of data, analysis or interpretation of data. JCD: acquisition of data, analysis or interpretation of data. IPB drafting/revising the manuscript, study concept or design, analysis or interpretation of data, study supervision, obtaining funding. GAN: drafting/revising the manuscript, study concept or design, contribution of vital reagents/tools/patients, study supervision, obtaining funding.

  • Funding This work was supported by the National Health and Medical Research Council of Australia (1004670, 511941) and a Peter Stearne grant from the MND Research Institute of Australia.

  • Competing interests MCK is the Editor-in-Chief of JNNP.

  • Ethics approval Ethics approval was granted by the Human Research Ethics Committee of the Sydney South West Area Health Service.

  • Provenance and peer review Not commissioned; externally peer reviewed.